Supporting Information Burford et al. 1.173/pnas.1339311 SI Materials and Methods β-arrestin Recruitment Assay. PathHunter human osteosarcoma cells (U2OS) expressing either μ-opioid receptors (U2OS- OPRM1) or δ-opioid receptors (U2OS-OPRD1), were grown in modified Eagle s medium containing 1% (vol/vol) FBS, 5 μg/ml G418, and 25 μg/ml hygromycin. Cells were grown to confluence in cell culture Nunc triple-layer flasks (Thermo Fisher Scientific), harvested with TrypLE Express, and resuspended in assay buffer (Hanks buffered salt solution plus 25 mm Hepes, 1 IU/mL penicillin, 1 μg/ml streptomycin,.5% BSA at 1 1 6 cells/ml). Compounds (2 nl of 1 final concentration in 1% DMSO) were added to white, nontreated 1,536-well plates (Corning) by acoustic dispense using an Echo-55 (Labcyte) from Echo-qualified 1,536-well source plates (Corning). Next, 1 μl of assay buffer (agonist-detection mode), assay buffer containing a low concentration ( EC 1 ) of orthosteric agonist [positive allosteric modulator (PAM)-detection mode], or assay buffer containing an EC 8 concentration of orthosteric agonist [antagonist/negative allosteric modulator (NAM)-detection mode] was added to assay plates. The orthosteric agonists used and their final concentrations are described in Results. Finally, 1 μl of cells (1, cells per well) in assay buffer was added to the wells to initiate the incubation period. Plates were lidded and incubated at room temperature for 9 min. Incubations were terminated by the addition of 1 μl of PathHunter reagent. One hour later, luminescence was detected using a Viewlux imaging plate reader (PerkinElmer). Additional characterization of certain μ-selective PAMs in the β-arrestin assay were performed essentially as described above using the various orthosteric agonist ligands and cell lines described in Results. Inhibition of Forskolin-Stimulated camp-accumulation Assays. CHO cells expressing recombinant human μ-opioid receptor (CHO-μ) were grown to confluence in F12 media containing 1% FBS, 1 IU/mL penicillin, 1 μg/ml streptomycin, and 4 μg/ml G418 in T-175 tissue culture flasks (Corning) and harvested with TrypLE Express. Cells were pelleted by centrifugation and resuspended in assay buffer at 6.67 1 5 cells per milliliter. Compounds (3 nl of 1 final concentration in 1% DMSO) were added to 1,536-well white, solid nontreated plates by acoustic dispense using an Echo-55. Next, 1.5 μl of assay buffer containing 1 mm 3-isobutyl-1-methylxanthine (IBMX) and 2 forskolin (final concentration, 1 μm), without (agonistdetection mode) or with (PAM-detection mode) 2 endomorphin-i (final concentration, 3 pm; an EC 1 concentration) were added to the plates. Finally, cells (1.5 μl per well) were added to begin the incubation. Plates were incubated at room temperature for 3 min, followed by the addition of Cisbio homogeneous time resolved fluoresecence resonance energy transfer (HTRF) dynamic camp detection reagent (1.5 μl of D2-labeled camp tracer in lysis buffer, followed by 1.5 μl of Eucryptate conjugated anti-camp antibody in lysis buffer). After a 1 h of incubation at room temperature, time-resolved fluorescence was detected on a Viewlux or Envision plate reader (PerkinElmer) with excitation at 337 nm and emission reads at 615 and 665 nm. The ratiometric data (665 nm read/615 nm read) 1, was then converted to camp (nanomoles per liter) based on a standard curve for camp (replacing the cell addition step) run at the same time and under identical conditions to the assay. Characterization of μ-selective PAMs in the CHO-μ camp assay, using curve-shift assays and probe-dependence assays, were performed as described above, using orthosteric agonists and modulators described in Results. [ 35 S]GTPγS Binding Assay. Membranes were diluted with 5 mm Tris HCl (ph 7.4) and preincubated with assay buffer containing GDP (1 volume membrane plus 2 volumes of 2 assay buffer) for 3 min at room temperature in a shaking water bath. Then, 15 μl of membrane/assay buffer mixture was added to wells containing 5 μl of drugs and [ 35 S]GTPγS [final concentrations: 2 mm Tris HCl (ph 7.4), 1 mm NaCl, 5 mm MgCl 2,.1 mm DTT, 3 1 μm GDP,.1 nm [ 35 S]GTPγS, 3 μm [D-Ala 2, N-MePhe 4, Gly-ol]-enkephalin (DAMGO) or morphine, 1 μm modulator or DMSO to achieve 2% DMSO final concentration and either 15 μg of membrane protein per well for C6μ cell membranes or 1 μg of membrane protein per well for mouse brain membranes]. After incubation for an additional 5 min at room temperature, samples were quickly filtered through glass-fiber filter mats using a Brandel cell harvester and rinsed five times with ice-cold wash buffer [5 mm Tris HCl (ph 7.4), 1 mm NaCl, 5mMMgCl 2 ]. Filter mats were dried, scintillation mixture was added, and radioactivity retained on the filters was counted in a Wallac MicroBeta (PerkinElmer). Data Analysis. Data from Fig. 2 were analyzed simultaneously with an allosteric ternary complex model in GraphPad Prism 5.1 (Dose response Special Allosteric EC 5 shift): Model : EC 5 = 1bLogEC 5 K b = 1bLogK b alpha = 1bLogalpha Antag = ð1 + B=K b Þ=ð1 + alpha B=K b Þ LogEC = LogðEC 5 AntagÞ Y = Bottom + ðtop BottomÞ=ð1 + 1bððLogEC XÞ HillSlopeÞÞ; where EC 5 is the concentration of orthosteric agonist that produces a half maximal response in the absence of modulator. K b is the equilibrium dissociation constant (molar) of modulator binding to its allosteric site. Alpha is the cooperativity factor. B is the concentration of allosteric modulator present. Top and Bottom are plateaus in the units of the Y axis. Burford et al. www.pnas.org/cgi/content/short/1339311 1of9
camp accumulation (nm) 15 1 5 EC 5 = 76 pm -13-12 -11-1 [endomorphin-i] log. M. Fig. S1. Effect of endomorphin-i on inhibition of 1 μm forskolin-stimulated camp accumulation in CHO-μ cells. Endomorphin-I induced a 17-fold reduction in forskolin-stimulated camp accumulation with an EC 5 of 76 pm [95% confidence interval (CI): 6 96 pm]. Data are represented as the means ± SEM of three experiments. Fig. S2. Effect of μ-pam, BMS-986121, on DAMGO-stimulated (A) and morphine-stimulated (B) [ 35 S]GTPγS binding in C6μ cell membranes. BMS-986121 (1 μm) resulted in a fourfold leftward shift in DAMGO potency. No significant agonist activity was detected for BMS-986121 (A). BMS-986121 (1 μm) increased morphine potency by 2.5-fold and increased the maximal effect (E max ) of morphine (B). Data are represented as the means ± SEM of three experiments. A 8 B 8 fmol bound/mg 6 4 2 vehicle fmol bound/mg 6 4 2 vehicle BMS-986123 1 2 3 4 [ 3 H] Diprenorphine, nm 1 2 3 4 [ 3 H] Diprenorphine, nm Fig. S3. Effect of the μ-pam,, and the silent allosteric modulator (SAM), BMS-986123, on [ 3 H]diprenorphine saturation binding in membranes from C6μ cells. (1 μm) (A) had no significant effect on [ 3 H]diprenorphine-binding affinity but induced a sixfold increase in the affinity of DAMGO in competition-binding studies (Fig. 4 and Table S1). BMS-986123 (1 μm) (B) produced a small ( twofold) but significant decrease in [ 3 H]diprenorphine affinity but had no significant effect on DAMGO affinity (Table S1). Data are represented as the means ± SEM of three to seven experiments. Burford et al. www.pnas.org/cgi/content/short/1339311 2of9
of [ 35 S]GTP S binding relative to 3 um DAMGO in C6mu membranes 1 8 6 4 2 vehicle [endomorphin-i] log. M. Fig. S4. Effect of μ-pam, (1 μm), on endomorphin-i stimulated [ 35 S]GTPγS binding in C6μ membranes. Endomorphin-I was a partial agonist in this system relative to DAMGO. (1 μm) increased the E max of endomorphin-i compared with DAMGO [from 65% (95% CI: 62 69%) to 84% (95% CI: 8 87%)] and decreased the EC 5 for endomorphin-i by twofold [from 37 nm (95% CI: 231 47 nm) to 143 nm (95% CI: 114 18 nm)]. Data are represented as the means ± SEM of three experiments. delta agonist mode delta PAM mode mu agonist mode mu PAM mode mu SAM mode % activity % activity % activity % activity % inhibition Substance U2OS-OPRD1 U2OS-OPRD1 U2OS-OPRM1 U2OS-OPRM1 U2OS-OPRM1 Analog 1 Analog 2 Analog 3 Analog 4 Fig. S5. (Continued) Burford et al. www.pnas.org/cgi/content/short/1339311 3of9
delta agonist mode delta PAM mode mu agonist mode mu PAM mode mu SAM mode % activity % activity % activity % activity % inhibition Substance U2OS-OPRD1 U2OS-OPRD1 U2OS-OPRM1 U2OS-OPRM1 U2OS-OPRM1 Analog 5 Analog 6 Analog 7 (BMS-986123) Analog 8 Analog 9 Fig. S5. (Continued) Burford et al. www.pnas.org/cgi/content/short/1339311 4of9
delta agonist mode delta PAM mode mu agonist mode mu PAM mode mu SAM mode % activity % activity % activity % activity % inhibition Substance U2OS-OPRD1 U2OS-OPRD1 U2OS-OPRM1 U2OS-OPRM1 U2OS-OPRM1 Analog 1 Analog 11 Analog 12 Analog 13 (BMS-986124) Analog 14 Analog 15 Fig. S5. Activity of and 15 analogs in a β-arrestin recruitment assay in U2OS-OPRD1 and U2OS-OPRM1 cells. Compounds were tested in U2OS- OPRD1 cells in agonist-detection mode (in the absence of leu-enkephalin) and PAM-detection mode (in the presence of an EC 1 concentration of leuenkephalin). Compounds were also tested in U2OS-OPRM1 cells in agonist-detection mode (in the absence of endomorphin-i) and PAM-detection mode (in the presence of an EC 1 concentration of endomorphin-i). Finally, compounds that exhibited no agonist or PAM activity in U2OS-OPRM1 cells were tested in SAMdetection mode (in the presence of an EC 2 concentration of endomorphin-i plus an EC 8 of the PAM ). Graphical curve fit data are representative of three combined experiments. EC 5 and E max values are represented in Table S2. For agonist-detection mode, % and 1% activity represent basal activity and an E max concentration of orthosteric agonist, respectively. For PAM-detection mode, % activity represents the response to a low ( EC 1 ) concentration of agonist alone, and 1% activity represents the response to an E max concentration of agonist. For SAM-detection mode, % inhibition represents the response to a low ( EC 2 ) concentration of endomorphin-i combined with an EC 8 concentration of the μ-pam ; 1% inhibition represents the response to a low ( EC 2 ) concentration of endomorphin-i alone. Burford et al. www.pnas.org/cgi/content/short/1339311 5of9
% activity EC 8 (3nM) endomorphin-i 1 8 6 4 2 BMS-986123 BMS-986124-7 -6-5 -4 [compound] log. M. Fig. S6. Effect of BMS-986123 and BMS-986124 on β-arrestin response to an EC 8 concentration of endomorphin-i (antagonist/nam-detection mode) in U2OS-OPRM1 cells. BMS-986123 and BMS-986124 had no significant effect on endomorphin-i mediated (3 nm) β-arrestin activity in U2OS-OPRM1 cells; 1% activity is normalized to the response to endomorphin-i (3 nm) alone; % activity represents basal activity. Data are represented as the means ± SEM of three experiments. A B 3 25 2 15 1 5 3 25 2 15 1 5 BMS-986124 C 3 25 2 15 1 5 + BMS-986124 Fig. S7. Effect of BMS-986124 on mediated PAM activity to DAMGO-stimulated [ 35 S]GTPγS binding in C6μ cell membranes. DAMGO-stimulated [ 35 S]GTPγS-binding potency in C6μ membranes was increased eightfold in the presence of the μ-pam (1 μm) (A). DAMGO-stimulated [ 35 S]GTPγSbinding potency was not affected by incubation with 5 μm BMS-986124 (B). Coincubation of BMS-986124 (5 μm) with (1 μm) resulted in a rightward shift in DAMGO potency compared with incubation with alone (C). The potency of DAMGO in the presence of both and BMS-986124 was shifted leftward by only twofold compared with DAMGO potency in the presence of the. These data suggest that BMS-986124 can antagonize the PAM effect. Shown are the combined means ± SEM data from three to seven separate assays, each performed in duplicate. EC 5 values are given in Results and Discussion and in Fig. 5B. Burford et al. www.pnas.org/cgi/content/short/1339311 6of9
A 3 25 2 15 1 5 BMS-986123 B 3 25 2 15 1 5 BMS-986124 Fig. S8. Effect of the SAMs, BMS-986123 and BMS-986124, on DAMGO-stimulated [ 35 S]GTPγS binding above basal activity in membranes from C6μ cells. DAMGO potency [EC 5 of 222 nm (95% CI: 179 274 nm)] was not significantly affected by BMS-986123 (A) [EC 5 of 321 nm (95% CI: 239 432 nm] or BMS- 986124 (B) [EC 5 of 223 nm (95% CI: 15 331 nm)]. The maximal stimulation by DAMGO [control max, 232% (95% CI: 223 242%)] was not affected by BMS- 986124 [243% (95% CI: 224 262%)]. Maximal stimulation was decreased slightly by BMS-986123 [26% (95% CI: 193 218%)]. The modulators did not significantly affect the basal values ( basal, 3.2 ±.2 fmol bound per milligram of protein). Shown are the combined data from three to seven separate assays, each performed in duplicate. A B relative to 3 um DAMGO 8 6 4 2 BMS-986123 relative to 3 um DAMGO 8 6 4 2 BMS-986124 [morphine] log. M. [morphine] log. M. Fig. S9. Effect of the SAMs BMS-986123 and BMS-986124 on morphine-stimulated [ 35 S]GTPγS binding in membranes from C6μ cells. The EC 5 of morphine to stimulate [ 35 S]GTPγS binding [11 nm (95% CI: 71 171 nm)] was not significantly affected by BMS-986123 (A) [14 nm (95% CI: 67 293 nm)] but was decreased by BMS-986124 (B) [245 nm (95% CI: 161 372 nm)]. The E max of morphine compared with DAMGO (3 μm) [control max, 42% (95% CI: 38 45%)] was increased to a small degree by BMS-986123 [62% (95% CI: 52 73%)] and BMS-986124 [58% (95% CI: 53 64%)]. Shown are the combined data from three to seven separate assays, each performed in duplicate. Burford et al. www.pnas.org/cgi/content/short/1339311 7of9
Fig. S1. Effect of μ-pam BMS-986121 on inhibition of forskolin-stimulated camp accumulation, mediated by different orthosteric agonists, in CHO-μ cells. BMS-986121 (1 μm) produced leftward shifts in agonist potency for each of the three orthosteric ligands used [endomorphin-i, fourfold (A); morphine, fivefold (B); and leu-enkephalin, fivefold (C)]. EC 5 values for the agonists at each BMS-986121 concentration are shown in the legend. Data represent the means ± SEM of three experiments. Table S1. Effect of the μ-pam and the SAM BMS-986123 on [ 3 H]diprenorphine saturation binding and DAMGO competition binding in membranes from C6μ cells Compound added [ 3 H]diprenorphine, K d nm, mean (95% CI) DAMGO, K i nm, mean (95% CI) Vehicle control in Tris HCl + sodium and GTP analogs.27 (.21.32) 34 (28-552) (1 μm) in Tris HCl + sodium and GTP analogs.35 (.18.51) 56 (41-76) (1 μm) in Tris HCl + sodium and GTP analogs.34 (.13.55) ND BMS-986123 (1 μm) in Tris HCl + sodium and GTP analogs.71 (.57.86) 27 (179-46) Vehicle control in Tris HCl buffer.22 (.2.46) 2.21 (1.51 3.23) (1 μm) in Tris HCl buffer.23 (.4.43).51 (.36.72) ND, not determined. Burford et al. www.pnas.org/cgi/content/short/1339311 8of9
Table S2. Structure activity relationship of and analogs tested in the β-arrestin recruitment assay in U2OS-OPRM1 cells, in PAM-detection mode Substance R1 R2 R3 R4 R5 β-arrestin PAMdetection mode, %E max β-arrestin PAMdetection mode, EC 5 (μm) Description β-arrestin SAM detection mode, K b (μm) H Cl OMe Br H 79 3 Full PAM (ref) Analog 1 H Nitro OMe Br H 53 2 Moderate PAM Analog 2 H Me OMe Br H 18 4 Weak PAM Analog 3 H OMe OMe Br H 16 6 Weak PAM Analog 4 Me H OMe Br H 2 9 Weak PAM Analog 5 H Cl OMe Nitro H 45 4 Moderate PAM Analog 6 H Br OMe Nitro H 45 2 Moderate PAM Analog 7 (BMS-986123) H Me OMe Nitro H SAM 1 Analog 8 H Br OMe OMe H 18 14 Weak PAM Analog 9 H Br OMe H H 44 7 Moderate PAM Analog 1 H H OMe H H 17 23 Weak PAM Analog 11 H Cl H Cl H 65 5 Strong PAM Analog 12 H Cl Br H H 26 6 Weak PAM Analog 13 (BMS-986124) H Cl Br H OMe SAM 2 Analog 14 H OMe Br H H 3 7 Weak PAM Analog 15 H OMe OMe H H 9 37 Weak PAM Compounds exhibiting PAM activity were described based on their efficacy as full, strong, moderate, or weak PAMs; % activity represents the response to alow( EC 1 ) concentration of endomorphin-i alone, and 1% activity represents the response to an E max concentration of endomorphin-i. The two compounds that showed no PAM activity were additionally tested in SAM-detection mode [inhibition of ( EC 8 ) response in the presence of a low concentration ( EC 2 ) of endomorphin-i], where the compounds were shown to inhibit the response (Fig. S5). Calculated K b values are provided from IC 5 values. Concentration response curves for the SAM compounds, BMS-986123 and BMS-986124, are shown in Fig. 5A. Data are represented as the means of three experiments., not active. Burford et al. www.pnas.org/cgi/content/short/1339311 9of9